I have a URL regex I use (and have used quite frequently). It does me well for finding various URL formats and http protocols. That said, I wouldn't be writing here if all was dandy in Dandyland.
I've encountered a hiccup that my current regex below is causing.
When searching a string for URLs, if a string consists of something like example...see it will treat it as a URL. There can be any number of periods, however it only pulls the last 3 characters after the last period.
Any ideas how to resolve this?
Example:
$string = "Here's a url, hello.com. But this...shouldn't show.";
$url_regex = "/((https?|ftp)\:\/\/)?([a-z0-9+!*(),;?&=\$_.-]+(\:[a-z0-9+!*(),;?&=\$_.-]+)?#)?([a-z0-9-.]*)\.([a-z]{2,3})(\:[0-9]{2,5})?(\/([a-z0-9+\$_\-~#\(\)\%]\.?)+)*\/?(\?[a-z+&\$_.-][a-z0-9;:#&#%=+\/\$_.-]*)?(#[a-z_.-][a-z0-9+\$_.-]*)?/i";
preg_match_all($url_regex, $string, $urls);
return $urls;
The problem here was that you had added a period within the allowed characters which meant there could be more than one consecutive periods. Also \b is important when you're dealing with inline searches.
\b((https?|ftp)\:\/\/)?([a-z0-9+!*(),;?&=\$_-]+(\:[a-z0-9+!*(),;?&=\$_-]+)?#)?([a-z0-9-]*)\.([a-z]+){2,3}(\:[0-9])?(\/([a-z0-9+\$_\-~#\(\)\%]?)+)*\/?(\?[a-z+&\$_-][a-z0-9;:#&#%=+\/\$_-]*)?(#[a-z_-][a-z0-9+\$_-]*)?\b
Debuggex Demo
Edit: Updated the answer to ignore matches like example.c
Following code solve your issue. I have test at my end.
$string = "Here's a url, hello.com. But this...shouldn't show.";
$url_regex = "/((https?|ftp)\:\/\/)?([a-z0-9+!*(),;?&=\$_.-]+(\:[a-z0-9+!*(),;?&=\$_.-]+)?#)?([a-z0-9-]+?)\.([a-z]{2,3})(\:[0-9]{2,5})?(\/([a-z0-9+\$_\-~#\(\)\%]\.?)+)*\/?(\?[a-z+&\$_.-][a-z0-9;:#&#%=+\/\$_.-]*)?(#[a-z_.-][a-z0-9+\$_.-]*)?/i";
preg_match_all($url_regex, $string, $urls);
Use https and http with urls in the string.
$string = "this is my website http://example.com and this is my friend website https://pqr.com etc, this...shouldn't show";
$regex = '/\b(https?|ftp|file):\/\/[-A-Z0-9+&##\/%?=~_|$!:,.;]*[A-Z0-9+&##\/%=~_|$]/i';
preg_match_all($regex, $string, $matches);
print_r($matches[0]);
Related
I'm not very good at regular expressions at all.
I've been using a lot of framework code to date, but I'm unable to find one that is able to match a URL like http://www.example.com/etcetc, but it is also is able to catch something like www.example.com/etcetc and example.com/etcetc.
For matching all kinds of URLs, the following code should work:
<?php
$regex = "((https?|ftp)://)?"; // SCHEME
$regex .= "([a-z0-9+!*(),;?&=$_.-]+(:[a-z0-9+!*(),;?&=$_.-]+)?#)?"; // User and Pass
$regex .= "([a-z0-9\-\.]*)\.(([a-z]{2,4})|([0-9]{1,3}\.([0-9]{1,3})\.([0-9]{1,3})))"; // Host or IP address
$regex .= "(:[0-9]{2,5})?"; // Port
$regex .= "(/([a-z0-9+$_%-]\.?)+)*/?"; // Path
$regex .= "(\?[a-z+&\$_.-][a-z0-9;:#&%=+/$_.-]*)?"; // GET Query
$regex .= "(#[a-z_.-][a-z0-9+$%_.-]*)?"; // Anchor
?>
Then, the correct way to check against the regex is as follows:
<?php
if(preg_match("~^$regex$~i", 'www.example.com/etcetc', $m))
var_dump($m);
if(preg_match("~^$regex$~i", 'http://www.example.com/etcetc', $m))
var_dump($m);
?>
Courtesy: Comments made by splattermania in the PHP manual: preg_match
RegEx Demo in regex101
This worked for me in all cases I had tested:
$url_pattern = '/((http|https)\:\/\/)?[a-zA-Z0-9\.\/\?\:#\-_=#]+\.([a-zA-Z0-9\&\.\/\?\:#\-_=#])*/';
Tests:
http://test.test-75.1474.stackoverflow.com/
https://www.stackoverflow.com
https://www.stackoverflow.com/
http://wwww.stackoverflow.com/
http://wwww.stackoverflow.com
http://test.test-75.1474.stackoverflow.com/
http://www.stackoverflow.com
http://www.stackoverflow.com/
stackoverflow.com/
stackoverflow.com
http://www.example.com/etcetc
www.example.com/etcetc
example.com/etcetc
user:pass#example.com/etcetc
example.com/etcetc?query=aasd
example.com/etcetc?query=aasd&dest=asds
http://stackoverflow.com/questions/6427530/regular-expression-pattern-to-match-url-with-or-without-http-www
http://stackoverflow.com/questions/6427530/regular-expression-pattern-to-match-url-with-or-without-http-www/
Every valid Internet URL has at least one dot, so the above pattern will simply try to find any at least two strings chained by a dot and has valid characters that URL may have.
Try this:
/^http:\/\/|(www\.)?[a-z0-9]+([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}(:[0-9]{1,5})?(\/.*)?$/
It works exactly like the people want.
It takes with or with out http://, https://, and www.
You can use a question mark after a regular expression to make it conditional so you would want to use:
http:\/\/(www\.)?
That will match anything that has either http://www. or http:// (with no www.)
You could just use a replace method to remove the above, thus getting you the domain. It depends on what you need the domain for.
Try something like this:
.*([\w-]+\.)+[a-z]{2,5}(/[\w-]+)*
Use:
/(https?://)?((?:(\w+-)*\w+)\.)+(?:[a-z]{2})(\/?\w?-?=?_?\??&?)+[\.]?([a-z0-9\?=&_\-%#])?/g
It matches something.com, http(s):// or www. It does not match other [something]:// URLs though, but for my purpose that's not necessary.
The regex matches e.g.:
http://foo.co.uk/
www.regex.com/foo.html?q=bar$some=thi-ng,regex
regex.foo.com/blog
You can try this:
r"(http[s]:\/\/)?([\w-]+\.)+([a-z]{2,5})(\/+\w+)? "
Selection:
may be start with http:// or https:// (optional)
anything (word) end with dot (.)
followed by 2 to 5 character [a-z]
followed by "/[anything]" (optional)
followed by space
Try this
$url_reg = /(ftp|https?):\/\/(\w+:?\w*#)?(\S+)(:[0-9]+)?(\/([\w#!:.?+=&%#!\/-])?)?/;
I have been using the following, which works for all my test cases, as well as fixes any issues where it would trigger at the end of a sentence preceded by a full-stop (end.), or where there were single character initials, such as 'C.C. Plumbing'.
The following regex contains multiple {2,}s, which means two or more matches of the previous pattern.
((http|https)\:\/\/)?[a-zA-Z0-9\.\/\?\:#\-_=#]{2,}\.([a-zA-Z0-9\&\.\/\?\:#\-_=#]){2,}
Matches URLs such as, but not limited to:
https://example.com
http://example.com
example.com
example.com/test
example.com?value=test
Does not match non-URLs such as, but not limited to:
C.C Plumber
A full-stop at the end of a sentence.
Single characters such as a.b or x.y
Please note: Due to the above, this will not match any single character URLs, such as: a.co, but it will match if it is preceded by a URL scheme, such as: http://a.co.
I was getting so many issues getting the answer from anubhava to work due to recent PHP allowing $ in strings and the preg match wasn't working.
Here is what I used:
// Regular expression
$re = '/((https?|ftp):\/\/)?([a-z0-9+!*(),;?&=.-]+(:[a-z0-9+!*(),;?&=.-]+)?#)?([a-z0-9\-\.]*)\.(([a-z]{2,4})|([0-9]{1,3}\.([0-9]{1,3})\.([0-9]{1,3})))(:[0-9]{2,5})?(\/([a-z0-9+%-]\.?)+)*\/?(\?[a-z+&$_.-][a-z0-9;:#&%=+\/.-]*)?(#[a-z_.-][a-z0-9+$%_.-]*)?/i';
// Match all
preg_match_all($re, $blob, $matches, PREG_SET_ORDER, 0);
// Print the entire match result
var_dump($matches);
// The first element of the array is the full match
This PHP Composer package URL highlight is doing a good job in PHP:
<?php
use VStelmakh\UrlHighlight\UrlHighlight;
$urlHighlight = new UrlHighlight();
$matches = $urlHighlight->getUrls($string);
?>
If it does not have to be regex, you could always use the validate filters that are in PHP.
filter_var('http://example.com', FILTER_VALIDATE_URL);
filter_var (mixed $variable [, int $filter = FILTER_DEFAULT [, mixed $options ]]);
Types of Filters
Validate Filters
Regex if you want to ensure a URL starts with HTTP/HTTPS:
https?:\/\/(www\.)?[-a-zA-Z0-9#:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()#:%_\+.~#?&//=]*)
If you do not require the HTTP protocol:
[-a-zA-Z0-9#:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()#:%_\+.~#?&//=]*)
I'm not very good at regular expressions at all.
I've been using a lot of framework code to date, but I'm unable to find one that is able to match a URL like http://www.example.com/etcetc, but it is also is able to catch something like www.example.com/etcetc and example.com/etcetc.
For matching all kinds of URLs, the following code should work:
<?php
$regex = "((https?|ftp)://)?"; // SCHEME
$regex .= "([a-z0-9+!*(),;?&=$_.-]+(:[a-z0-9+!*(),;?&=$_.-]+)?#)?"; // User and Pass
$regex .= "([a-z0-9\-\.]*)\.(([a-z]{2,4})|([0-9]{1,3}\.([0-9]{1,3})\.([0-9]{1,3})))"; // Host or IP address
$regex .= "(:[0-9]{2,5})?"; // Port
$regex .= "(/([a-z0-9+$_%-]\.?)+)*/?"; // Path
$regex .= "(\?[a-z+&\$_.-][a-z0-9;:#&%=+/$_.-]*)?"; // GET Query
$regex .= "(#[a-z_.-][a-z0-9+$%_.-]*)?"; // Anchor
?>
Then, the correct way to check against the regex is as follows:
<?php
if(preg_match("~^$regex$~i", 'www.example.com/etcetc', $m))
var_dump($m);
if(preg_match("~^$regex$~i", 'http://www.example.com/etcetc', $m))
var_dump($m);
?>
Courtesy: Comments made by splattermania in the PHP manual: preg_match
RegEx Demo in regex101
This worked for me in all cases I had tested:
$url_pattern = '/((http|https)\:\/\/)?[a-zA-Z0-9\.\/\?\:#\-_=#]+\.([a-zA-Z0-9\&\.\/\?\:#\-_=#])*/';
Tests:
http://test.test-75.1474.stackoverflow.com/
https://www.stackoverflow.com
https://www.stackoverflow.com/
http://wwww.stackoverflow.com/
http://wwww.stackoverflow.com
http://test.test-75.1474.stackoverflow.com/
http://www.stackoverflow.com
http://www.stackoverflow.com/
stackoverflow.com/
stackoverflow.com
http://www.example.com/etcetc
www.example.com/etcetc
example.com/etcetc
user:pass#example.com/etcetc
example.com/etcetc?query=aasd
example.com/etcetc?query=aasd&dest=asds
http://stackoverflow.com/questions/6427530/regular-expression-pattern-to-match-url-with-or-without-http-www
http://stackoverflow.com/questions/6427530/regular-expression-pattern-to-match-url-with-or-without-http-www/
Every valid Internet URL has at least one dot, so the above pattern will simply try to find any at least two strings chained by a dot and has valid characters that URL may have.
Try this:
/^http:\/\/|(www\.)?[a-z0-9]+([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}(:[0-9]{1,5})?(\/.*)?$/
It works exactly like the people want.
It takes with or with out http://, https://, and www.
You can use a question mark after a regular expression to make it conditional so you would want to use:
http:\/\/(www\.)?
That will match anything that has either http://www. or http:// (with no www.)
You could just use a replace method to remove the above, thus getting you the domain. It depends on what you need the domain for.
Try something like this:
.*([\w-]+\.)+[a-z]{2,5}(/[\w-]+)*
Use:
/(https?://)?((?:(\w+-)*\w+)\.)+(?:[a-z]{2})(\/?\w?-?=?_?\??&?)+[\.]?([a-z0-9\?=&_\-%#])?/g
It matches something.com, http(s):// or www. It does not match other [something]:// URLs though, but for my purpose that's not necessary.
The regex matches e.g.:
http://foo.co.uk/
www.regex.com/foo.html?q=bar$some=thi-ng,regex
regex.foo.com/blog
You can try this:
r"(http[s]:\/\/)?([\w-]+\.)+([a-z]{2,5})(\/+\w+)? "
Selection:
may be start with http:// or https:// (optional)
anything (word) end with dot (.)
followed by 2 to 5 character [a-z]
followed by "/[anything]" (optional)
followed by space
Try this
$url_reg = /(ftp|https?):\/\/(\w+:?\w*#)?(\S+)(:[0-9]+)?(\/([\w#!:.?+=&%#!\/-])?)?/;
I have been using the following, which works for all my test cases, as well as fixes any issues where it would trigger at the end of a sentence preceded by a full-stop (end.), or where there were single character initials, such as 'C.C. Plumbing'.
The following regex contains multiple {2,}s, which means two or more matches of the previous pattern.
((http|https)\:\/\/)?[a-zA-Z0-9\.\/\?\:#\-_=#]{2,}\.([a-zA-Z0-9\&\.\/\?\:#\-_=#]){2,}
Matches URLs such as, but not limited to:
https://example.com
http://example.com
example.com
example.com/test
example.com?value=test
Does not match non-URLs such as, but not limited to:
C.C Plumber
A full-stop at the end of a sentence.
Single characters such as a.b or x.y
Please note: Due to the above, this will not match any single character URLs, such as: a.co, but it will match if it is preceded by a URL scheme, such as: http://a.co.
I was getting so many issues getting the answer from anubhava to work due to recent PHP allowing $ in strings and the preg match wasn't working.
Here is what I used:
// Regular expression
$re = '/((https?|ftp):\/\/)?([a-z0-9+!*(),;?&=.-]+(:[a-z0-9+!*(),;?&=.-]+)?#)?([a-z0-9\-\.]*)\.(([a-z]{2,4})|([0-9]{1,3}\.([0-9]{1,3})\.([0-9]{1,3})))(:[0-9]{2,5})?(\/([a-z0-9+%-]\.?)+)*\/?(\?[a-z+&$_.-][a-z0-9;:#&%=+\/.-]*)?(#[a-z_.-][a-z0-9+$%_.-]*)?/i';
// Match all
preg_match_all($re, $blob, $matches, PREG_SET_ORDER, 0);
// Print the entire match result
var_dump($matches);
// The first element of the array is the full match
This PHP Composer package URL highlight is doing a good job in PHP:
<?php
use VStelmakh\UrlHighlight\UrlHighlight;
$urlHighlight = new UrlHighlight();
$matches = $urlHighlight->getUrls($string);
?>
If it does not have to be regex, you could always use the validate filters that are in PHP.
filter_var('http://example.com', FILTER_VALIDATE_URL);
filter_var (mixed $variable [, int $filter = FILTER_DEFAULT [, mixed $options ]]);
Types of Filters
Validate Filters
Regex if you want to ensure a URL starts with HTTP/HTTPS:
https?:\/\/(www\.)?[-a-zA-Z0-9#:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()#:%_\+.~#?&//=]*)
If you do not require the HTTP protocol:
[-a-zA-Z0-9#:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()#:%_\+.~#?&//=]*)
I am attempting to change a string occurance e.g. http://www.bbc.co.uk/ so that it appears inside a html link e.g. http://www.bbc.co.uk
however for some reason my regex conversion does not work. Can someone please point me in the correct direction?
$text = "I love this website http://www.bbc.co.uk/";
$x = preg_replace("#[a-z]+://[^<>\s]+[[a-z0-9]/]#i", "\\0", $text);
var_dump($x);
outputs I love this website http://www.bbc.co.uk/ (No html link)
Your weird character class is at fault:
[[a-z0-9]/]
Double square brackets are for POSIX character classes like [[:digit:]].
You meant to write just:
[a-z0-9/]
It is because you regex is giving you a match (in fact it's really not even close to giving you a match as you are not accepting periods in the domain name at all). Try something like this:
$pattern = '#https?://.*\b#i';
$replace = '$0';
$x = preg_replace($pattern, $replace, $text);
Note that I am not actually trying to validate the URL format here, so I just accept anything like http():// up to the next word boundary. It didn't seem as if you were going for a true URL validation regex anyway (i.e. validating there is at least one ., that the TLD component has 2-6 characters, etc.), so I just figure I would give you the simplest pattern that would match.
Use this:
$x = preg_replace('#http://[?=&a-z0-9._/-]+#i', '<a target="_blank" href="$0">$0</a>', $text);
for example i have data response from here:
http://www.facebook.com/ajax/shares/view/?target_fbid=410558838979218&__a=1
there is a pattern that looks like this:
data-hovercard=\"\/ajax\/hovercard\/hovercard.php?id=655581307\">
how can i parse it with the preg_match_all() in PHP?
I know i need complex regular expression, but i dont have a clue how to write one for such pattern in the text.
Thanks for help
UPD:
the following code does give the id:
$str = 'hovercard.php?id=655581307';
preg_match_all('/[0-9]{9}/', $str , $matches);
print_r($matches);
BUT
this one doesnt
$url = 'http://www.facebook.com/ajax/shares/view/?target_fbid=410558838979218&__a=1';
$html = file_get_contents($url);
preg_match_all('/[0-9]{9}/', $html, $matches);
print_r($matches);
This gets a bit messy due to the backslashes escaping stuff, but to match exactly that string this call to preg_match_all() should work:
preg_match_all('#(data-hovercard=\\\\"\\\\/ajax\\\\/hovercard\\\\/hovercard.php\?id=[0-9]+\\\\">)#', $str, $matches);
That will give you the whole string you posted in $matches. However, if you only want the numbers from id you can add extra parenthesis around that like so:
preg_match_all('#(data-hovercard=\\\\"\\\\/ajax\\\\/hovercard\\\\/hovercard.php\?id=([0-9]+)\\\\">)#', $str, $matches);
And the numbers will appear individually in $matches (similarly, you can remove the parenthesis that wraps the whole regexp to stop matching the whole string).
Update:
And now I see the question is updated. If your new example fails it's because there are no sequence of 9 digits in the data you get. When I try myself I simply get a response that says I need to log in, so maybe your matching issues is in fact due to you not getting the data you expect? Try dumping $html to see if what you are looking for is in fact in there.
I'm not very good at regular expressions at all.
I've been using a lot of framework code to date, but I'm unable to find one that is able to match a URL like http://www.example.com/etcetc, but it is also is able to catch something like www.example.com/etcetc and example.com/etcetc.
For matching all kinds of URLs, the following code should work:
<?php
$regex = "((https?|ftp)://)?"; // SCHEME
$regex .= "([a-z0-9+!*(),;?&=$_.-]+(:[a-z0-9+!*(),;?&=$_.-]+)?#)?"; // User and Pass
$regex .= "([a-z0-9\-\.]*)\.(([a-z]{2,4})|([0-9]{1,3}\.([0-9]{1,3})\.([0-9]{1,3})))"; // Host or IP address
$regex .= "(:[0-9]{2,5})?"; // Port
$regex .= "(/([a-z0-9+$_%-]\.?)+)*/?"; // Path
$regex .= "(\?[a-z+&\$_.-][a-z0-9;:#&%=+/$_.-]*)?"; // GET Query
$regex .= "(#[a-z_.-][a-z0-9+$%_.-]*)?"; // Anchor
?>
Then, the correct way to check against the regex is as follows:
<?php
if(preg_match("~^$regex$~i", 'www.example.com/etcetc', $m))
var_dump($m);
if(preg_match("~^$regex$~i", 'http://www.example.com/etcetc', $m))
var_dump($m);
?>
Courtesy: Comments made by splattermania in the PHP manual: preg_match
RegEx Demo in regex101
This worked for me in all cases I had tested:
$url_pattern = '/((http|https)\:\/\/)?[a-zA-Z0-9\.\/\?\:#\-_=#]+\.([a-zA-Z0-9\&\.\/\?\:#\-_=#])*/';
Tests:
http://test.test-75.1474.stackoverflow.com/
https://www.stackoverflow.com
https://www.stackoverflow.com/
http://wwww.stackoverflow.com/
http://wwww.stackoverflow.com
http://test.test-75.1474.stackoverflow.com/
http://www.stackoverflow.com
http://www.stackoverflow.com/
stackoverflow.com/
stackoverflow.com
http://www.example.com/etcetc
www.example.com/etcetc
example.com/etcetc
user:pass#example.com/etcetc
example.com/etcetc?query=aasd
example.com/etcetc?query=aasd&dest=asds
http://stackoverflow.com/questions/6427530/regular-expression-pattern-to-match-url-with-or-without-http-www
http://stackoverflow.com/questions/6427530/regular-expression-pattern-to-match-url-with-or-without-http-www/
Every valid Internet URL has at least one dot, so the above pattern will simply try to find any at least two strings chained by a dot and has valid characters that URL may have.
Try this:
/^http:\/\/|(www\.)?[a-z0-9]+([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}(:[0-9]{1,5})?(\/.*)?$/
It works exactly like the people want.
It takes with or with out http://, https://, and www.
You can use a question mark after a regular expression to make it conditional so you would want to use:
http:\/\/(www\.)?
That will match anything that has either http://www. or http:// (with no www.)
You could just use a replace method to remove the above, thus getting you the domain. It depends on what you need the domain for.
Try something like this:
.*([\w-]+\.)+[a-z]{2,5}(/[\w-]+)*
Use:
/(https?://)?((?:(\w+-)*\w+)\.)+(?:[a-z]{2})(\/?\w?-?=?_?\??&?)+[\.]?([a-z0-9\?=&_\-%#])?/g
It matches something.com, http(s):// or www. It does not match other [something]:// URLs though, but for my purpose that's not necessary.
The regex matches e.g.:
http://foo.co.uk/
www.regex.com/foo.html?q=bar$some=thi-ng,regex
regex.foo.com/blog
You can try this:
r"(http[s]:\/\/)?([\w-]+\.)+([a-z]{2,5})(\/+\w+)? "
Selection:
may be start with http:// or https:// (optional)
anything (word) end with dot (.)
followed by 2 to 5 character [a-z]
followed by "/[anything]" (optional)
followed by space
Try this
$url_reg = /(ftp|https?):\/\/(\w+:?\w*#)?(\S+)(:[0-9]+)?(\/([\w#!:.?+=&%#!\/-])?)?/;
I have been using the following, which works for all my test cases, as well as fixes any issues where it would trigger at the end of a sentence preceded by a full-stop (end.), or where there were single character initials, such as 'C.C. Plumbing'.
The following regex contains multiple {2,}s, which means two or more matches of the previous pattern.
((http|https)\:\/\/)?[a-zA-Z0-9\.\/\?\:#\-_=#]{2,}\.([a-zA-Z0-9\&\.\/\?\:#\-_=#]){2,}
Matches URLs such as, but not limited to:
https://example.com
http://example.com
example.com
example.com/test
example.com?value=test
Does not match non-URLs such as, but not limited to:
C.C Plumber
A full-stop at the end of a sentence.
Single characters such as a.b or x.y
Please note: Due to the above, this will not match any single character URLs, such as: a.co, but it will match if it is preceded by a URL scheme, such as: http://a.co.
I was getting so many issues getting the answer from anubhava to work due to recent PHP allowing $ in strings and the preg match wasn't working.
Here is what I used:
// Regular expression
$re = '/((https?|ftp):\/\/)?([a-z0-9+!*(),;?&=.-]+(:[a-z0-9+!*(),;?&=.-]+)?#)?([a-z0-9\-\.]*)\.(([a-z]{2,4})|([0-9]{1,3}\.([0-9]{1,3})\.([0-9]{1,3})))(:[0-9]{2,5})?(\/([a-z0-9+%-]\.?)+)*\/?(\?[a-z+&$_.-][a-z0-9;:#&%=+\/.-]*)?(#[a-z_.-][a-z0-9+$%_.-]*)?/i';
// Match all
preg_match_all($re, $blob, $matches, PREG_SET_ORDER, 0);
// Print the entire match result
var_dump($matches);
// The first element of the array is the full match
This PHP Composer package URL highlight is doing a good job in PHP:
<?php
use VStelmakh\UrlHighlight\UrlHighlight;
$urlHighlight = new UrlHighlight();
$matches = $urlHighlight->getUrls($string);
?>
If it does not have to be regex, you could always use the validate filters that are in PHP.
filter_var('http://example.com', FILTER_VALIDATE_URL);
filter_var (mixed $variable [, int $filter = FILTER_DEFAULT [, mixed $options ]]);
Types of Filters
Validate Filters
Regex if you want to ensure a URL starts with HTTP/HTTPS:
https?:\/\/(www\.)?[-a-zA-Z0-9#:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()#:%_\+.~#?&//=]*)
If you do not require the HTTP protocol:
[-a-zA-Z0-9#:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()#:%_\+.~#?&//=]*)